Radial Engine Cooling

[Pursuivant]

Something screwy that I've noticed is that identical engines with similar cooling setups sometimes behave in very different fashions. For example, it's very hard to get the Wright R-1820-40 Cyclone on the B-239 Buffalo to overheat, but much easier to get it to overheat on the Curtis-Wright CW-21 Demon.

For liquid-cooled engines I can understand if there's more variation in the time that it takes for the engine to overheat or cool down, since you're dealing with things like radiator surface area and heat transfer efficiency.

For radials, it seems like there shouldn't be so much variation, though, since you've basically got a big round engine and the size and shape of the cooling flaps is more or less defined by the engine's diameter.